[Question] Correct way of hyper parameter optimization for new algorithm?

[tyler-ingebrand]

❓ Question

Hello,

I am writing a new algorithm based on the sb3 implementation of PPO. However, I need to run hyper parameter optimization since I changed some parts of the algorithm and the defaults no longer work for me. What is the best way to test a new algorithm, which implements BasePolicy, using this package? Do I need to just clone the repo and modify utils.py, hyperparams_opt.py, etc? Or is there a cleaner way?

Thank you for your time

Checklist

• [X] I have checked that there is no similar issue in the repo
• [X] I have read the SB3 documentation
• [X] I have read the RL Zoo documentation
• [X] If code there is, it is minimal and working
• [X] If code there is, it is formatted using the markdown code blocks for both code and stack traces.

[araffin]

Hello, yes, a fork would be the way to go. If you just want to add/replace an algorithm, you can take a look at https://rl-baselines3-zoo.readthedocs.io/en/master/guide/sbx.html